SPATIAL AUTHENTICATION SCHEME FOR ONLINE TRANSACTIONS

• Main Authors: Che-Wei Hsu, 許哲瑋
• Other Authors: 歐陽彥杰
• Format: Others
• Language: en_US
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/85093108324605913276

碩士 === 國立中興大學 === 電機工程學系所 === 100 === Humans are not machines. Humans are organisms with emotions and conscious. The different states of humans make us error-prone and weaker than machines at performing logical tasks. Humans are the weakest link in a network security system and human-machine interaction occurs during the authentication phase. The chain of a security system is only as good as the weakest link. Strengthening the authentication phase can strengthen the overall security of a system. There is a trade-off between usability and security in existing authentication methods. Longer and random passwords used to authenticate users increase security but diminish usability and vice versa for shorter and common passwords. The problem with existing authentication methods lie in the password itself. Passwords need to be long and random, changed on a regular basis but also easy to remember. We proposed an innovative authentication scheme which combined the concept of space and image. Passwords were selected from a picture through mouse clicks. In terms of usability, more memorable points arose from using pictures compared to random text due to the amount of context pictures had in association with users. In terms of security, our proposed method had a larger pool of passwords to choose from which yielded in more password combinations. We constructed a prototype of the authentication system and produced theoretical results through analytical models. Results showed that our proposed method had at one point 796 times more password combinations compared to existing alphanumeric passwords. Our authentication method displayed improvements in both usability and security. This research developed a platform for further improvements to current authentication methods to help strengthen the overall security and performance of network security systems.